In view of the lack of controlled data on the effectiveness of inpatient family intervention practice, it is proposed to do a controlled study of the relative effectiveness (as measured by 18-month community follow-up of hospitalization with family intervention, as currently practiced at PWC and other inpatient facilities, for schizophrenic disorder and major affective disorder patients in need of hospital care and for whom both treatments are judged clinically feasible. In addition, we aim to determine patient characteristic associated with differential benefits from these two types of treatment. The sample will include 80 schizophrenic disorder and 50 major affective disorder patients, who at the time of admission will be randomly assigned to one of two groups on the same ward: an "experimental" group, receiving the multimodal hospital treatment with inpatient family intervention and a comparison group, receiving the more generally practices mutimodal hospital treatment (which does not include inpatient family intervention). Measures of change will be obtained at the time of hospital discharge and at six- and 18-month follow-up using the Psychiatric Evaluation Form, the Global Assessment Scale, the Strauss-Carpenter Levels of Function Scale, the Social Adjustment Scale, the Strauss-Carpenter Levels of Function Scale, the Social Adjustment Scale (SR), the Role Performance Treatment Scale, the Patient's Self Evaluation of Current Status, the Family Member's Evaluation of Current Status of Patient, and the Treatment and Medication Compliance Data Scale. In addition, the Family Attitude Scale, Goals of Family Therapy Rating Scale, and a modified Goal Attainment Scale will be used to study family change and individual patient outcome. The long-range objective of this research project is to assess the relative cost-effectiveness of inpatient family intervention in the prevention of post-discharge replapse and rehospitalization.